Abstract:

Process Mining has become an established field of research addressing a variety of problems using abroad range of methods. In this talk, I will show what Natural Language Processing (NLP) can bring to the table. Specifically, I will focus on the problem of anomaly detection. Anomaly detection in process mining aims to recognize outlying or unexpected behavior in event logs for purposes such as the removal of noise and identification of conformance violations. Existing techniques for this task are primarily frequency-based, arguing that behavior is anomalous because it is uncommon. However, such techniques ignore the semantics of recorded events and, therefore, do not take the meaning of potential anomalies into consideration.

In this talk, I will show how to overcome this limitation, arguing that anomalies can be recognized when process behavior does not make sense. Building on this, I will show how NLP can also be leveraged for other relevant problems in Process Mining. I will close with an outlook how NLP could even help to automatically improve business processes.